INDEX

Explanations

specific sequences of words, potentially related to a storyline or narrative

oai_token-act-pair · gpt-3.5-turbo

elements related to personal stories and individual experiences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

artney

-0.63

abal

-0.58

farious

-0.58

undreds

-0.55

luster

-0.54

senal

-0.53

 mosqu

-0.51

 undermin

-0.50

iership

-0.50

zyme

-0.49

POSITIVE LOGITS

tho

0.57

 however

0.54

 ital

0.53

 grate

0.49

 certainly

0.49

 reprint

0.47

 nudity

0.46

 Reply

0.45

 dont

0.45

 wiki

0.45

Activations Density 1.128%

specific sequences of words, potentially related to a storyline or narrative

elements related to personal stories and individual experiences

No Comments

No Known Activations